Instructions to use Zaynoid/JSL-Granite-CP with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Zaynoid/JSL-Granite-CP with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Zaynoid/JSL-Granite-CP")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Zaynoid/JSL-Granite-CP")
model = AutoModelForCausalLM.from_pretrained("Zaynoid/JSL-Granite-CP")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Zaynoid/JSL-Granite-CP with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Zaynoid/JSL-Granite-CP"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Zaynoid/JSL-Granite-CP",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Zaynoid/JSL-Granite-CP

SGLang

How to use Zaynoid/JSL-Granite-CP with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "Zaynoid/JSL-Granite-CP" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Zaynoid/JSL-Granite-CP",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "Zaynoid/JSL-Granite-CP" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Zaynoid/JSL-Granite-CP",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Zaynoid/JSL-Granite-CP with Docker Model Runner:
```
docker model run hf.co/Zaynoid/JSL-Granite-CP
```

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

JSL-Granite-CP

A fine-tune of ibm-granite/granite-4.0-h-tiny (Mamba-2 hybrid MoE, ~7B total / ~1B active) that authors Spark NLP ContextualParser (CP) rules for PHI de-identification.

Given a clinical text and a target entity type, it emits a CP rule as JSON — a regex plus context gates (prefix/suffix words, contextLength, contextException), or a dictionary, or null (abstain) when no rule applies.

Results — clean held-out test (700 rows, zero training leakage)

model	F1	hallucinations
base `granite-4.0-h-tiny`	88.33	288
JSL-Granite-CP (this model)	93.97	39

Scored with the project's shipped reward logic (grpo_reward / offline simulate_cp — the rule is executed against the text and matched spans are compared to gold). Test rows share 0 examples with the training data.

Training

QLoRA (bitsandbytes 4-bit nf4, LoRA r=16 / α=32, target_modules="all-linear") on 5,029 reward-verified CP-rule examples (balanced across entities, weak entities oversampled), 2 epochs, then merged back to bf16.

Usage

Use the project system prompt (from to_chat.py). Serve with vLLM:

vllm serve Zaynoid/JSL-Granite-CP --max-model-len 8192 --trust-remote-code

User message format:

Text: <clinical text>

Target entity: <ENTITY>

Return the ContextualParser rule as JSON.

Output: {"cp_rule": { ... }} or {"cp_rule": null}.

Downloads last month: 38

Safetensors

Model size

7B params

Tensor type

BF16

Model tree for Zaynoid/JSL-Granite-CP

Base model

ibm-granite/granite-4.0-h-tiny

Finetuned

(10)

this model